Reliability and Software metrics Done by Tayeb El

Reliability and Software metrics Done by: Tayeb El Alaoui Software Engineering II Course

Outline q. Introduction § Team presentation § Context presentation § Subject presentation q. What is Reliability ? ( Definition from IEEE )

Outline q. Difference between Errors, Faults and Failures. q. Software Metrics for Reliability § Requirements Reliability Metrics. § Design and code Reliability Metrics. § Testing Reliability Metrics. q. Conclusion

Introduction q. Team presentation § Amine Mestari (Senior student) § Tayeb El Alaoui (Junior student)

Introduction q. Context presentation § To complete the requirements of our Software Engineering II course. § To improve my communication skills.

Introduction q. Subject presentation § Reliability = correctness. (For most software developers) § A better way: To go through all stages of Software lifecycle: Requirements, design and code, and Testing.

What is Reliability ? q. Definition from IEEE: “The ability of a system or component to perform its required functions under stated conditions for a specified period of time. "

What is Reliability ? q. Three activities: § Error prevention § Fault detection and removal § Measurements to maximize Reliability

Errors, Faults and Failures q. Error: programmer action or omission that results in a fault q. Fault: software defect that causes a failure. q. Failure: the unacceptable departure of a program operation.

Software Failure rate

Software Metrics for Reliability q. The goal: To help improve the reliability of a software. q. How ? To identify some potential errors from different phases of software development.

Requirements Reliability Metrics q. Use: To assess individual specification statements and the vocabulary used. q. How? ARM software (Automated Requirement Measurement)

Requirements Reliability Metrics (From SATC and NASA) q q q q Lines of Text Imperatives Continuances Directives Weak Phrases Incomplete Options

Example of Analysis study

Design and code Reliability Metrics q. Complexity (Cyclomatic) q. Size q. Modularity q. Weighted Method per Class (WMC)

Testing Reliability Metrics q. Two approaches: § Evaluation of the test plan. § Evaluation of the number of errors in the code and the rate of finding / fixing them.

Evaluation of the test plan q. Use of multiple test cases q. A test case contains several functions q. Each function must be at least tested once.

Conclusion q. Existence of Metrics to evaluate Reliability. At each phase of the development lifecycle. q. The cost benefits are: factor of 14.

References q Musa, J. D. , A. Iannino and K. Okumoto, Software Reliability: Measurement, Prediction, Application, Professional Edition: Software Engineering Series, Mc. Graw–Hill, New York, NY. , 1990. q Triantafyllos, G. , S. Vassiliadis and W. Kobrosly, "On the Prediction of Computer Implementation Faults Via Static Error Prediction Models, " Journal of Systems and Software, Vol. 28, No. 2, February 1995, pp. 129 -142. q Rosenberg, L. , and Hammer, T. , Metrics for Quality Assurance and Risk Assessment, Proc. Eleventh International Software Quality Week, San Francisco, CA, 1998. q Hammer, T. , Rosenberg, L. , Huffman, L. , Hyatt, L. , Measuring Requirements Testing in Proc. International Conference on Software Engineering (Boston MA, May 1997) IEEE Computer Society Press.